gx-100: lsusb, lspci...?

Post by bad sector
I'd like to nail down how my system sees my
(Roland) Boss-gx-100 guitar effects board.
It's connected via a USB cable but neither
of the commands in the subject header lists
it regardless of whether it's powered on or not.

It should be in "lsusb", but it doesn't need
to have an identifier, if nobody bothered to inform
the keeper of the informal USB.ids file.

0582:0289 <no text available to go here>

The official USB.org file is unavailable. This is
to protect prototype equipment ("lab specimens") from
exposure to the world. If Linux could access the official
list, the quality of "lsusb" would be "authoritative".

The unofficial list, relies on somebody to make an
identification, and then a string is assigned as
an identifier with the new code.

http://www.linux-usb.org/usb.ids

0582 Roland Corp.

It's not in that section.

*******

So what I do, is try to find an INF file in a Windows driver, and
work out what the two hex numbers should be.

https://static.roland.com/assets/media/zip/gx100_w1011d_v100DL.zip

[Roland.NTamd64.10]
;; Windows10
%DriverDeviceDesc%=DriverInstall, USB\VID_0582&PID_0289 ; GX-100

So that is the VID and PID value that should be detected.

lsusb should have a 0582:0289 entry, but unless it is listed
in a usb.ids file, there won't be a text string next to those
two numbers in the lsusb output.

*******

When you're in Windows, you can use USBTreeView to see the
"thing" appear when you plug it in. It's only going to show
in Windows Device Manager (devmgmt.msc) if you've installed
the driver. But this tool, like lsusb, should show the device
even without a driver. It likely forms endpoints and then
without a driver, will not do much else.

https://www.uwe-sieber.de/usbtreeview_e.html

Download latest release:

x64:
https://www.uwe-sieber.de/files/UsbTreeView_x64.zip (~420KB)

Win32:
https://www.uwe-sieber.de/files/UsbTreeView_Win32.zip (~390KB)

Paul

bad sector

2024-02-20 23:52:36 UTC

It should be in "lsusb", but it doesn't need
to have an identifier, if nobody bothered to inform
the keeper of the informal USB.ids file.
0582:0289 <no text available to go here>
The official USB.org file is unavailable. This is
to protect prototype equipment ("lab specimens") from
exposure to the world. If Linux could access the official
list, the quality of "lsusb" would be "authoritative".
The unofficial list, relies on somebody to make an
identification, and then a string is assigned as
an identifier with the new code.
http://www.linux-usb.org/usb.ids
0582 Roland Corp.
It's not in that section.
*******
So what I do, is try to find an INF file in a Windows driver, and
work out what the two hex numbers should be.
https://static.roland.com/assets/media/zip/gx100_w1011d_v100DL.zip
[Roland.NTamd64.10]
;; Windows10
%DriverDeviceDesc%=DriverInstall, USB\VID_0582&PID_0289 ; GX-100
So that is the VID and PID value that should be detected.
lsusb should have a 0582:0289 entry, but unless it is listed
in a usb.ids file, there won't be a text string next to those
two numbers in the lsusb output.
*******
When you're in Windows, you can use USBTreeView to see the
"thing" appear when you plug it in. It's only going to show
in Windows Device Manager (devmgmt.msc) if you've installed
the driver. But this tool, like lsusb, should show the device
even without a driver. It likely forms endpoints and then
without a driver, will not do much else.
https://www.uwe-sieber.de/usbtreeview_e.html
https://www.uwe-sieber.de/files/UsbTreeView_x64.zip (~420KB)
https://www.uwe-sieber.de/files/UsbTreeView_Win32.zip (~390KB)
Paul

Thank you, it *is* there when it's turned ON. Yesterday it wasn't
sjhhowing up at all. In a vBox hosted w10 it showed up as "Boss GX-100"
but the last reintall in vBox that I made doesn't show it anymore. I'll
try another run at it after making me a step-by-step to get at least
similar if not iudentical results every time :-)

# lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 001 Device 004: ID 13b1:003e Linksys AE6000 802.11a/b/g/n/ac
Wireless Adapter [MediaTek MT7610U]
Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 007: ID 058f:6387 Alcor Micro Corp. Flash Drive
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 002: ID 2f68:0082 Hoksi Technology DURGOD Taurus K320
Bus 005 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 006 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 007 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 007 Device 002: ID 046d:c062 Logitech, Inc. M-UAS144 [LS1 Laser Mouse]
Bus 008 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 009 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

Turn unit ON:

/home/u3 # lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 001 Device 004: ID 13b1:003e Linksys AE6000 802.11a/b/g/n/ac
Wireless Adapter [MediaTek MT7610U]
Bus 001 Device 005: ID 0582:0289 Roland Corp. GX-100
Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 007: ID 058f:6387 Alcor Micro Corp. Flash Drive
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 002: ID 2f68:0082 Hoksi Technology DURGOD Taurus K320
Bus 005 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 006 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 007 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 007 Device 002: ID 046d:c062 Logitech, Inc. M-UAS144 [LS1 Laser Mouse]
Bus 008 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 009 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

Paul

2024-02-21 08:39:10 UTC

It should be in "lsusb", but it doesn't need
to have an identifier, if nobody bothered to inform
the keeper of the informal USB.ids file.
     0582:0289   <no text available to go here>
The official USB.org file is unavailable. This is
to protect prototype equipment ("lab specimens") from
exposure to the world. If Linux could access the official
list, the quality of "lsusb" would be "authoritative".
The unofficial list, relies on somebody to make an
identification, and then a string is assigned as
an identifier with the new code.
http://www.linux-usb.org/usb.ids
0582 Roland Corp.
It's not in that section.
*******
So what I do, is try to find an INF file in a Windows driver, and
work out what the two hex numbers should be.
https://static.roland.com/assets/media/zip/gx100_w1011d_v100DL.zip
[Roland.NTamd64.10]
;; Windows10
%DriverDeviceDesc%=DriverInstall, USB\VID_0582&PID_0289 ; GX-100
So that is the VID and PID value that should be detected.
lsusb should have a 0582:0289 entry, but unless it is listed
in a usb.ids file, there won't be a text string next to those
two numbers in the lsusb output.
*******
When you're in Windows, you can use USBTreeView to see the
"thing" appear when you plug it in. It's only going to show
in Windows Device Manager (devmgmt.msc) if you've installed
the driver. But this tool, like lsusb, should show the device
even without a driver. It likely forms endpoints and then
without a driver, will not do much else.
https://www.uwe-sieber.de/usbtreeview_e.html
    https://www.uwe-sieber.de/files/UsbTreeView_x64.zip       (~420KB)
    https://www.uwe-sieber.de/files/UsbTreeView_Win32.zip     (~390KB)
   Paul

Thank you, it *is* there when it's turned ON. Yesterday it wasn't sjhhowing up at all. In a vBox hosted w10 it showed up as "Boss GX-100" but the last reintall in vBox that I made doesn't show it anymore. I'll try another run at it after making me a step-by-step to get at least similar if not iudentical results every time :-)
# lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 001 Device 004: ID 13b1:003e Linksys AE6000 802.11a/b/g/n/ac Wireless Adapter [MediaTek MT7610U]
Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 007: ID 058f:6387 Alcor Micro Corp. Flash Drive
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 002: ID 2f68:0082 Hoksi Technology DURGOD Taurus K320
Bus 005 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 006 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 007 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 007 Device 002: ID 046d:c062 Logitech, Inc. M-UAS144 [LS1 Laser Mouse]
Bus 008 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 009 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
/home/u3 # lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 001 Device 004: ID 13b1:003e Linksys AE6000 802.11a/b/g/n/ac Wireless Adapter [MediaTek MT7610U]
Bus 001 Device 005: ID 0582:0289 Roland Corp. GX-100
Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 007: ID 058f:6387 Alcor Micro Corp. Flash Drive
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 002: ID 2f68:0082 Hoksi Technology DURGOD Taurus K320
Bus 005 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 006 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 007 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 007 Device 002: ID 046d:c062 Logitech, Inc. M-UAS144 [LS1 Laser Mouse]
Bus 008 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 009 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

Install VirtualBox Additions, to a newly installed Guest virtual machine,
if you expect the hardware passthru to work. It's probably necessary
for this purpose.

Then, go to the Settings and add the item to the passthru list.
When the Guest starts up, the device will "disappear" from the Host,
and be "newly discovered hardware" in the Guest. And vice versa,
on shutdown.

Paul

bad sector

2024-02-22 00:44:46 UTC

It should be in "lsusb", but it doesn't need
to have an identifier, if nobody bothered to inform
the keeper of the informal USB.ids file.
     0582:0289   <no text available to go here>
The official USB.org file is unavailable. This is
to protect prototype equipment ("lab specimens") from
exposure to the world. If Linux could access the official
list, the quality of "lsusb" would be "authoritative".
The unofficial list, relies on somebody to make an
identification, and then a string is assigned as
an identifier with the new code.
http://www.linux-usb.org/usb.ids
0582 Roland Corp.
It's not in that section.
*******
So what I do, is try to find an INF file in a Windows driver, and
work out what the two hex numbers should be.
https://static.roland.com/assets/media/zip/gx100_w1011d_v100DL.zip
[Roland.NTamd64.10]
;; Windows10
%DriverDeviceDesc%=DriverInstall, USB\VID_0582&PID_0289 ; GX-100
So that is the VID and PID value that should be detected.
lsusb should have a 0582:0289 entry, but unless it is listed
in a usb.ids file, there won't be a text string next to those
two numbers in the lsusb output.
*******
When you're in Windows, you can use USBTreeView to see the
"thing" appear when you plug it in. It's only going to show
in Windows Device Manager (devmgmt.msc) if you've installed
the driver. But this tool, like lsusb, should show the device
even without a driver. It likely forms endpoints and then
without a driver, will not do much else.
https://www.uwe-sieber.de/usbtreeview_e.html
    https://www.uwe-sieber.de/files/UsbTreeView_x64.zip       (~420KB)
    https://www.uwe-sieber.de/files/UsbTreeView_Win32.zip     (~390KB)
   Paul

Install VirtualBox Additions, to a newly installed Guest virtual machine,
if you expect the hardware passthru to work. It's probably necessary
for this purpose.
Then, go to the Settings and add the item to the passthru list.
When the Guest starts up, the device will "disappear" from the Host,
and be "newly discovered hardware" in the Guest. And vice versa,
on shutdown.
Paul

Top-Left is initials, Extension pack is installed.
There is no mention of guest-tools (installed).

Bottom-Left is the intial USB filter in settings
showing the correct numbers for a GX-100

Top-Right is where it hits the fan, all the USB
checkboxes are ghosted, lsusb on top of it still
shows the device under Linux 'jurisdiction'.

Bottom-Right is a weird situation. So far during my
installs I clicked on 'hit any key to load from CD/DVD'
at the very start (maybe I shouldn't). A couple of times
I forgot and that got me going in a somewhat different
sequence of dialogs, including one where the installer
has to be re-pointed to the same iso image. One of these
dialogs, in this image, shows a transluscent screen in
the top-right about mouse and kybrd integration. Somehow
THIS little bit of transluscent info is NOWHERE to be
seen in my 'normal?' installs. Does thsi have any value
for troubleshooting?

David W. Hodgins

2024-02-22 02:26:34 UTC

Post by bad sector
80-90% of my vBox installations run into this problem and it's likely
something I'm doing wrong or not doing at all.
Tonight I tried on Suse Slowroll 7.0.14-2.1.
Just about everything virtualbox is installed
including "virtualbox guest tools 7.0.14-2.1"
which is supposedly "VirtualBox guest addition
tools" I cannot find any guest tools 'iso'
7.014-2.1 anywhere.

The guest additions must be installed in the guest. It's not needed in the
host.

In the host ...
- virtualbox must be installed
- matching version of the extension pack downloaded from
https://download.virtualbox.org/virtualbox/7.0.14/Oracle_VM_VirtualBox_Extension_Pack-7.0.14.vbox-extpack

- Virtualbox kernel modules built for the host's running kernel version and
the virtualbox version (vboxdrv, vboxnetadp, and vboxnetflt kernel modules).

In the guest, the guest additions must be installed. That can come either
from linux distro as a package, or in a running guest use the "Devices" menu
entry to insert the guest additions iso. If it hasn't been previously downloaded,
you'll be asked whether it should be downloaded (from virtualbox.org).

Regards, Dave Hodgins

bad sector

2024-02-22 21:54:23 UTC

For the moment I just did a freshie on Suse Slowroll,
none of the above package names avail but I have these:

python3 virtualbox
virtualbox-devel
virtualbox-guest-desktop-icons
virtualbox-guest-tools
virtualbox-kmp-default
(this one is upgradable but dependencies fail)
virtualbox-qt
virtualbox-vnc

Post by David W. Hodgins
In the guest, the guest additions must be installed. That can come either
from linux distro as a package, or in a running guest use the "Devices" menu
entry to insert the guest additions iso. If it hasn't been previously downloaded,
you'll be asked whether it should be downloaded (from virtualbox.org).
Regards, Dave Hodgins

Thanks, that's exactly how it goes but the
download fails, at least I presume the error
message is one of failure.

"
Downloading Guest Additions
Name: VBoxGuestAdditions
During certificate downloading:
Unknown reason
......................100%
"

Does the Geuest Additions iso HAVE to
be exactly the same version #? Maybe I
could just DL a next-best-one from somewhere?

David W. Hodgins

2024-02-23 01:18:04 UTC

On Thu, 22 Feb 2024 16:54:23 -0500, bad sector <***@_invalid.net> wrote:
<snip>

Thanks, that's exactly how it goes but the
download fails, at least I presume the error
message is one of failure.
"
Downloading Guest Additions
Name: VBoxGuestAdditions
Unknown reason
......................100%
"
Does the Geuest Additions iso HAVE to
be exactly the same version #? Maybe I
could just DL a next-best-one from somewhere?

It has to match the version of the kernel running on the host.

I'm running virtualbox-7.0.14-1.mga9 on the host.

When I select the devices menu entry to insert the guest additions iso, and
then select to download it, I get ...

"The VirtualBox Guest Additions disk image file has been successfully downloaded from https://download.virtualbox.org/virtualbox/7.0.14/VBoxGuestAdditions_7.0.14.iso and saved locally as /home/dave/.VirtualBox/VBoxGuestAdditions_7.0.14.iso.
Do you wish to register this disk image file and insert it into the virtual optical drive?"

As I'm using kde plasma, after inserting the disk image file, I use the "Disks
and Devices" systray entry to mount and open it, which opened it in dolphin.

Opening konsole, and the using "su -" to switch to root, the iso has been mounted
at "/run/media/dave/VBox_GAs_7.0.14". Note the version number matches the vb
version on the host. As root, I run
=============
# cd /run/media/dave/VBox_GAs_7.0.14
# ls -l|grep -i linux
-r-xr-xr-x 1 dave dave 6306247 Jan 15 09:03 VBoxLinuxAdditions.run*
[***@x9v VBox_GAs_7.0.14]# ./VBoxLinuxAdditions.run
Verifying archive integrity... 100% MD5 checksums are OK. All good.
Uncompressing VirtualBox 7.0.14 Guest Additions for Linux 100%
VirtualBox Guest Additions installer
This system appears to have a version of the VirtualBox Guest Additions
already installed. If it is part of the operating system and kept up-to-date,
there is most likely no need to replace it. If it is not up-to-date, you
should get a notification when you start the system. If you wish to replace
it with this version, please do not continue with this installation now, but
instead remove the current version first, following the instructions for the
operating system.

If your system simply has the remains of a version of the Additions you could
not remove you should probably continue now, and these will be removed during
installation.

Do you wish to continue? [yes or no]
yes
Copying additional installer modules ...
Installing additional modules ...
VirtualBox Guest Additions: Starting.
VirtualBox Guest Additions: Setting up modules
VirtualBox Guest Additions: Building the VirtualBox Guest Additions kernel
modules. This may take a while.
VirtualBox Guest Additions: To build modules for other installed kernels, run
VirtualBox Guest Additions: /sbin/rcvboxadd quicksetup <version>
VirtualBox Guest Additions: or
VirtualBox Guest Additions: /sbin/rcvboxadd quicksetup all
VirtualBox Guest Additions: Building the modules for kernel
6.6.14-desktop-2.mga9.
dracut: Executing: /usr/bin/dracut -f --kver 6.6.14-desktop-2.mga9
dracut: dracut module 'mksh' will not be installed, because command 'mksh' could not be found!
<snip>
dracut: dracut: using auto-determined compression method 'gzip'
dracut: *** Creating initramfs image file '/boot/initrd-6.6.14-desktop-2.mga9.img' done ***
VirtualBox Guest Additions: Running kernel modules will not be replaced until
the system is restarted or 'rcvboxadd reload' triggered
VirtualBox Guest Additions: reloading kernel modules and services
VirtualBox Guest Additions: cannot reload kernel modules: one or more module(s)
is still in use
VirtualBox Guest Additions: kernel modules and services were not reloaded
The log file /var/log/vboxadd-setup.log may contain further information.
=============

Note that it uses dkms to build the modules for the kernel running in the guest,
so the dkms package must be installed, and the portion of the kernel source used
to compile kernel modules

In mageia those packages are called dkms, dkms-minimal, and for the kernel
source kernel-server-devel (or kernel-desktop-devel, etc., depending on the
flavor of the kernel that's running).

You can see in the above output, I already had the guest additions installed,
hence the kernel modules were already running.

As the guest is also running Mageia 9, I had installed in the guest ...
$ rpm -qa|grep virtualbox
virtualbox-guest-additions-7.0.14-1.mga9

I did the above changes in a snapshot for the vb guest so I can easily undo
the changes.

Regards, Dave Hodgins

Paul

2024-02-22 04:06:06 UTC

It should be in "lsusb", but it doesn't need
to have an identifier, if nobody bothered to inform
the keeper of the informal USB.ids file.
      0582:0289   <no text available to go here>
The official USB.org file is unavailable. This is
to protect prototype equipment ("lab specimens") from
exposure to the world. If Linux could access the official
list, the quality of "lsusb" would be "authoritative".
The unofficial list, relies on somebody to make an
identification, and then a string is assigned as
an identifier with the new code.
http://www.linux-usb.org/usb.ids
0582 Roland Corp.
It's not in that section.
*******
So what I do, is try to find an INF file in a Windows driver, and
work out what the two hex numbers should be.
https://static.roland.com/assets/media/zip/gx100_w1011d_v100DL.zip
[Roland.NTamd64.10]
;; Windows10
%DriverDeviceDesc%=DriverInstall, USB\VID_0582&PID_0289 ; GX-100
So that is the VID and PID value that should be detected.
lsusb should have a 0582:0289 entry, but unless it is listed
in a usb.ids file, there won't be a text string next to those
two numbers in the lsusb output.
*******
When you're in Windows, you can use USBTreeView to see the
"thing" appear when you plug it in. It's only going to show
in Windows Device Manager (devmgmt.msc) if you've installed
the driver. But this tool, like lsusb, should show the device
even without a driver. It likely forms endpoints and then
without a driver, will not do much else.
https://www.uwe-sieber.de/usbtreeview_e.html
     https://www.uwe-sieber.de/files/UsbTreeView_x64.zip       (~420KB)
     https://www.uwe-sieber.de/files/UsbTreeView_Win32.zip     (~390KB)
    Paul

Install VirtualBox Additions, to a newly installed Guest virtual machine,
if you expect the hardware passthru to work. It's probably necessary
for this purpose.
Then, go to the Settings and add the item to the passthru list.
When the Guest starts up, the device will "disappear" from the Host,
and be "newly discovered hardware" in the Guest. And vice versa,
on shutdown.
Paul

80-90% of my vBox installations run into this problem and it's likely something I'm doing wrong or not doing at all.
Tonight I tried on Suse Slowroll 7.0.14-2.1.
Just about everything virtualbox is installed
including "virtualbox guest tools 7.0.14-2.1"
which is supposedly "VirtualBox guest addition
tools" I cannot find any guest tools 'iso'
7.014-2.1 anywhere.
I uploaded a screen shot with 4 captures on my desktop
https://i.imgur.com/ZPWTblA.png
Top-Left is initials, Extension pack is installed.
There is no mention of guest-tools (installed).
Bottom-Left is the intial USB filter in settings
showing the correct numbers for a GX-100
Top-Right is where it hits the fan, all the USB
checkboxes are ghosted, lsusb on top of it still
shows the device under Linux 'jurisdiction'.
Bottom-Right is a weird situation. So far during my
installs I clicked on 'hit any key to load from CD/DVD'
at the very start (maybe I shouldn't). A couple of times
I forgot and that got me going in a somewhat different
sequence of dialogs, including one where the installer
has to be re-pointed to the same iso image. One of these
dialogs, in this image, shows a transluscent screen in
the top-right about mouse and kybrd integration. Somehow
THIS little bit of transluscent info is NOWHERE to be
seen in my 'normal?' installs. Does thsi have any value
for troubleshooting?

The only additional idea that comes to mind, is watch
that your Filter terms are not too tight. Like, say you
insist it is on Port 002 in the filter, then you come back
with a coffee in hand, and idly move the USB cable from
Port 002 to Port 003, without thinking of the Filter.

If other devices were passed through (say, for example,
a test USB stick used to test passthru), then check that
the filter for the GX-100 hasn't become too tight for the job.

If other items are passing through OK, the configuration
files needed, must already be present. That leaves the Filter.
(Or, a device malfunction, which is unlikely.)

Paul

bad sector

2024-02-22 22:01:09 UTC

It should be in "lsusb", but it doesn't need
to have an identifier, if nobody bothered to inform
the keeper of the informal USB.ids file.
      0582:0289   <no text available to go here>
The official USB.org file is unavailable. This is
to protect prototype equipment ("lab specimens") from
exposure to the world. If Linux could access the official
list, the quality of "lsusb" would be "authoritative".
The unofficial list, relies on somebody to make an
identification, and then a string is assigned as
an identifier with the new code.
http://www.linux-usb.org/usb.ids
0582 Roland Corp.
It's not in that section.
*******
So what I do, is try to find an INF file in a Windows driver, and
work out what the two hex numbers should be.
https://static.roland.com/assets/media/zip/gx100_w1011d_v100DL.zip
[Roland.NTamd64.10]
;; Windows10
%DriverDeviceDesc%=DriverInstall, USB\VID_0582&PID_0289 ; GX-100
So that is the VID and PID value that should be detected.
lsusb should have a 0582:0289 entry, but unless it is listed
in a usb.ids file, there won't be a text string next to those
two numbers in the lsusb output.
*******
When you're in Windows, you can use USBTreeView to see the
"thing" appear when you plug it in. It's only going to show
in Windows Device Manager (devmgmt.msc) if you've installed
the driver. But this tool, like lsusb, should show the device
even without a driver. It likely forms endpoints and then
without a driver, will not do much else.
https://www.uwe-sieber.de/usbtreeview_e.html
     https://www.uwe-sieber.de/files/UsbTreeView_x64.zip       (~420KB)
     https://www.uwe-sieber.de/files/UsbTreeView_Win32.zip     (~390KB)
    Paul

Install VirtualBox Additions, to a newly installed Guest virtual machine,
if you expect the hardware passthru to work. It's probably necessary
for this purpose.
Then, go to the Settings and add the item to the passthru list.
When the Guest starts up, the device will "disappear" from the Host,
and be "newly discovered hardware" in the Guest. And vice versa,
on shutdown.
Paul

The only additional idea that comes to mind, is watch
that your Filter terms are not too tight. Like, say you
insist it is on Port 002 in the filter, then you come back
with a coffee in hand, and idly move the USB cable from
Port 002 to Port 003, without thinking of the Filter.
If other devices were passed through (say, for example,
a test USB stick used to test passthru), then check that
the filter for the GX-100 hasn't become too tight for the job.
If other items are passing through OK, the configuration
files needed, must already be present. That leaves the Filter.
(Or, a device malfunction, which is unlikely.)

The filter fills itself if I try to add the board while it is plugged-in
and turned on, saves a lot of footwork. Last night it was all so
screwed up that it was asking me for partition numbers to install to and
finally gave up the ghost with some error about not being able to locate
a certain file. Deleted everything vbox from system and started anew,
MUCH faster than before and the previously mentioned 'mouse/keybord
integration' annunciation was there also until "I" closed it. Everything
seems to be OK except for the usb/GuestAdditions iso.

bad sector

2024-02-23 01:27:03 UTC

The system sees it, vBox sees, it but w10 does not.

vBox is presently uninstallable on Tumbleweed.

Oracle say to use the GuestAdditions supplied
with the package but OpenSuse packages do not
include it.

I found a Debian package @

https://download.opensuse.org/repositories/home:/pmdpalma:/bucket/Debian_12/all/virtualbox-guest-additions-iso_7.0.14-1_all.deb

and converted it to rpm and installed it in host
OpenSuse-Slowroll, then executed its installer in
the w10 guest.

Still no cigar! The usb listings remain ghosted and
'lsusb' continues to list GX-100 meaning that Slowroll
is keeping jurisdiction.

David W. Hodgins

2024-02-23 01:42:48 UTC

The system sees it, vBox sees, it but w10 does not.
vBox is presently uninstallable on Tumbleweed.
Oracle say to use the GuestAdditions supplied
with the package but OpenSuse packages do not
include it.
https://download.opensuse.org/repositories/home:/pmdpalma:/bucket/Debian_12/all/virtualbox-guest-additions-iso_7.0.14-1_all.deb
and converted it to rpm and installed it in host
OpenSuse-Slowroll, then executed its installer in
the w10 guest.
Still no cigar! The usb listings remain ghosted and
'lsusb' continues to list GX-100 meaning that Slowroll
is keeping jurisdiction.

The guest additions must be installed in the guest. As it's a windows guest,
it needs a .exe file to install it, not a linux script or executable.

In the iso image, it has ...
# tree -ifa|grep exe
./cert/VBoxCertUtil.exe
./NT3x/VBoxAddInstallNt3x.exe
./NT3x/VBoxControl.exe
./NT3x/VBoxService.exe
./OS2/VBoxControl.exe
./OS2/VBoxOs2AdditionsInstall.exe
./OS2/VBoxReplaceDll.exe
./OS2/VBoxService.exe
./VBoxWindowsAdditions-amd64.exe
./VBoxWindowsAdditions.exe
./VBoxWindowsAdditions-x86.exe

Regards, Dave Hodgins

bad sector

2024-02-23 04:56:43 UTC

The system sees it, vBox sees, it but w10 does not.
vBox is presently uninstallable on Tumbleweed.
Oracle say to use the GuestAdditions supplied
with the package but OpenSuse packages do not
include it.
https://download.opensuse.org/repositories/home:/pmdpalma:/bucket/Debian_12/all/virtualbox-guest-additions-iso_7.0.14-1_all.deb
and converted it to rpm and installed it in host
OpenSuse-Slowroll, then executed its installer in
the w10 guest.
Still no cigar! The usb listings remain ghosted and
'lsusb' continues to list GX-100 meaning that Slowroll
is keeping jurisdiction.

The guest additions must be installed in the guest. As it's a windows guest,
it needs a .exe file to install it, not a linux script or executable.
In the iso image, it has ...
# tree -ifa|grep exe
./cert/VBoxCertUtil.exe
./NT3x/VBoxAddInstallNt3x.exe
./NT3x/VBoxControl.exe
./NT3x/VBoxService.exe
./OS2/VBoxControl.exe
./OS2/VBoxOs2AdditionsInstall.exe
./OS2/VBoxReplaceDll.exe
./OS2/VBoxService.exe
./VBoxWindowsAdditions-amd64.exe

This *is* the one I had execute, no cigar. When I plug-in the unit,
Linux flashes the 'new sink detected gx-100' or words to that effect but
the usb list at the bottom of the w10 sccreen remains all ghosted. I
noticed that to personalize any settings I'd need to 'activate', maybe
this applies to usb passthrough as well. I remember at least 2 previous
dummy installs that I did NOT activate either but on which the USB
passthrough worked. I guess I could activate with the laptop key but
that's no garantee that it'll work.

Post by David W. Hodgins
./VBoxWindowsAdditions.exe
./VBoxWindowsAdditions-x86.exe

Paul

2024-02-23 08:03:30 UTC

The system sees it, vBox sees, it but w10 does not.
vBox is presently uninstallable on Tumbleweed.
Oracle say to use the GuestAdditions supplied
with the package but OpenSuse packages do not
include it.
https://download.opensuse.org/repositories/home:/pmdpalma:/bucket/Debian_12/all/virtualbox-guest-additions-iso_7.0.14-1_all.deb
and converted it to rpm and installed it in host
OpenSuse-Slowroll, then executed its installer in
the w10 guest.
Still no cigar! The usb listings remain ghosted and
'lsusb' continues to list GX-100 meaning that Slowroll
is keeping jurisdiction.

The guest additions must be installed in the guest. As it's a windows guest,
it needs a .exe file to install it, not a linux script or executable.
In the iso image, it has ...
# tree -ifa|grep exe
./cert/VBoxCertUtil.exe
./NT3x/VBoxAddInstallNt3x.exe
./NT3x/VBoxControl.exe
./NT3x/VBoxService.exe
./OS2/VBoxControl.exe
./OS2/VBoxOs2AdditionsInstall.exe
./OS2/VBoxReplaceDll.exe
./OS2/VBoxService.exe
./VBoxWindowsAdditions-amd64.exe

This *is* the one I had execute, no cigar. When I plug-in the unit, Linux flashes the 'new sink detected gx-100' or words to that effect but the usb list at the bottom of the w10 sccreen remains all ghosted. I noticed that to personalize any settings I'd need to 'activate', maybe this applies to usb passthrough as well. I remember at least 2 previous dummy installs that I did NOT activate either but on which the USB passthrough worked. I guess I could activate with the laptop key but that's no garantee that it'll work.

Post by David W. Hodgins
./VBoxWindowsAdditions.exe
./VBoxWindowsAdditions-x86.exe

In Host:

sudo usermod -a -G vboxusers bullwinkle # Add "vboxusers" to my username "bullwinkle" account

cat /etc/group # Bullwinkle should show up in the vboxusers entry
groups # My logged in account bullwinkle, should show the group
# but as long as the previous command looks OK, carry on.

Reboot. I suspect some Vbox daemon needs to
do something, to finish the job. You can't just
exit the Virtualbox panel and enter again,
and have the USB subsystem picked up. It
takes a reboot.

It's a good thing, on my Linux multiboot disk, there
were two OSes with VBox installed, one working,
one broken. I did the above to the broken one and
it's working now.

Paul

David W. Hodgins

2024-02-23 13:45:49 UTC

On Fri, 23 Feb 2024 03:03:30 -0500, Paul <***@needed.invalid> wrote:
<snip>

Post by Paul
sudo usermod -a -G vboxusers bullwinkle # Add "vboxusers" to my username "bullwinkle" account

Ahh. I forgot that part. It's been a while since I created a new installation.

Post by Paul
cat /etc/group # Bullwinkle should show up in the vboxusers entry
groups # My logged in account bullwinkle, should show the group
# but as long as the previous command looks OK, carry on.
Reboot. I suspect some Vbox daemon needs to
do something, to finish the job. You can't just
exit the Virtualbox panel and enter again,
and have the USB subsystem picked up. It
takes a reboot.

Reboot is not needed. Just logout/in for the user's group changes to take effect.

Post by Paul
It's a good thing, on my Linux multiboot disk, there
were two OSes with VBox installed, one working,
one broken. I did the above to the broken one and
it's working now.

Glad it's working.

Regards, Dave Hodgins

bad sector

2024-02-23 22:33:22 UTC

The system sees it, vBox sees, it but w10 does not.
vBox is presently uninstallable on Tumbleweed.
Oracle say to use the GuestAdditions supplied
with the package but OpenSuse packages do not
include it.
https://download.opensuse.org/repositories/home:/pmdpalma:/bucket/Debian_12/all/virtualbox-guest-additions-iso_7.0.14-1_all.deb
and converted it to rpm and installed it in host
OpenSuse-Slowroll, then executed its installer in
the w10 guest.
Still no cigar! The usb listings remain ghosted and
'lsusb' continues to list GX-100 meaning that Slowroll
is keeping jurisdiction.

The guest additions must be installed in the guest. As it's a windows guest,
it needs a .exe file to install it, not a linux script or executable.
In the iso image, it has ...
# tree -ifa|grep exe
./cert/VBoxCertUtil.exe
./NT3x/VBoxAddInstallNt3x.exe
./NT3x/VBoxControl.exe
./NT3x/VBoxService.exe
./OS2/VBoxControl.exe
./OS2/VBoxOs2AdditionsInstall.exe
./OS2/VBoxReplaceDll.exe
./OS2/VBoxService.exe
./VBoxWindowsAdditions-amd64.exe

This *is* the one I had execute, no cigar. When I plug-in the unit, Linux flashes the 'new sink detected gx-100' or words to that effect but the usb list at the bottom of the w10 sccreen remains all ghosted. I noticed that to personalize any settings I'd need to 'activate', maybe this applies to usb passthrough as well. I remember at least 2 previous dummy installs that I did NOT activate either but on which the USB passthrough worked. I guess I could activate with the laptop key but that's no garantee that it'll work.

Post by David W. Hodgins
./VBoxWindowsAdditions.exe
./VBoxWindowsAdditions-x86.exe

sudo usermod -a -G vboxusers bullwinkle # Add "vboxusers" to my username "bullwinkle" account
cat /etc/group # Bullwinkle should show up in the vboxusers entry
groups # My logged in account bullwinkle, should show the group
# but as long as the previous command looks OK, carry on.
Reboot. I suspect some Vbox daemon needs to
do something, to finish the job. You can't just
exit the Virtualbox panel and enter again,
and have the USB subsystem picked up. It
takes a reboot.
It's a good thing, on my Linux multiboot disk, there
were two OSes with VBox installed, one working,
one broken. I did the above to the broken one and
it's working now.

I think I would not even have been able to install
a win10 guest if I were not in group vboxusers. Even
the installer warns about that so I've been careful
to comply. What does vbox log, is there any way to
find and id the snag that prevents usb passthrough?

David W. Hodgins

2024-02-23 23:39:14 UTC

Post by bad sector
I think I would not even have been able to install
a win10 guest if I were not in group vboxusers. Even
the installer warns about that so I've been careful
to comply. What does vbox log, is there any way to
find and id the snag that prevents usb passthrough?

Double check the settings for the guest to ensure it has usb 2 or 3 enabled
matching what the host sees. Usb 2 is for ohci/ehci, while 3 is for xhci.

Regards, Dave Hodgins

bad sector

2024-02-24 03:44:20 UTC

Double check the settings for the guest to ensure it has usb 2 or 3 enabled
matching what the host sees. Usb 2 is for ohci/ehci, while 3 is for xhci.

The gx-100 was made in 2022-2023 so I presume it's all usb-3. My
Crosshair-IV Formula mobo has 2 usb3 and 6 usb3 ports, both types are
enabl;ed in BIOS. If I plug the gx-100 into a blue usb-3 port then
lsusb will NOT show it. Go figure. If I plug it into a usb-2 port then
it shows up.

The vBox host usb setup is a bit crooked but if I select usb-2 then it
will autodetect the device and auto-fill the the id numbers. Here the
device should NOT show up in the filter with usb-3 seleted because the
host cannot see it then. Anyway, I select usb-2 and the filter is
enabled though I don't see the point of this filter at all, all detected
usb devices show up under the guest window regardless of what's in this
filter, including the gx-100 if usb-2 is selected in the host.

If I set 3 in host then the guest w10 Device-Manager last entry also
shows USB 3, if I set 2 in the host then Device-Manager shows 2. The
very few times I got it to work I never had to tinker in the guest
Device-Manager. If I set usb-2 in the host, filter or no filter, then
the gx-100 shows up in the strip just below the w10 window BUT GHOSTED
and the bubble for it shows all the right numbers plus that it is
"unavailable".

David W. Hodgins

2024-02-24 04:17:00 UTC

Double check the settings for the guest to ensure it has usb 2 or 3 enabled
matching what the host sees. Usb 2 is for ohci/ehci, while 3 is for xhci.

The gx-100 was made in 2022-2023 so I presume it's all usb-3. My
Crosshair-IV Formula mobo has 2 usb3 and 6 usb3 ports, both types are
enabl;ed in BIOS. If I plug the gx-100 into a blue usb-3 port then
lsusb will NOT show it. Go figure. If I plug it into a usb-2 port then
it shows up.
The vBox host usb setup is a bit crooked but if I select usb-2 then it
will autodetect the device and auto-fill the the id numbers. Here the
device should NOT show up in the filter with usb-3 seleted because the
host cannot see it then. Anyway, I select usb-2 and the filter is
enabled though I don't see the point of this filter at all, all detected
usb devices show up under the guest window regardless of what's in this
filter, including the gx-100 if usb-2 is selected in the host.
If I set 3 in host then the guest w10 Device-Manager last entry also
shows USB 3, if I set 2 in the host then Device-Manager shows 2. The
very few times I got it to work I never had to tinker in the guest
Device-Manager. If I set usb-2 in the host, filter or no filter, then
the gx-100 shows up in the strip just below the w10 window BUT GHOSTED
and the bubble for it shows all the right numbers plus that it is
"unavailable".

Is the extension pack installed in the host? If that's been answered before,
I've forgotten.

https://download.virtualbox.org/virtualbox/7.0.14/Oracle_VM_VirtualBox_Extension_Pack-7.0.14.vbox-extpack

Regards, Dave Hodgins

bad sector

2024-02-24 04:46:54 UTC

Double check the settings for the guest to ensure it has usb 2 or 3 enabled
matching what the host sees. Usb 2 is for ohci/ehci, while 3 is for xhci.

The gx-100 was made in 2022-2023 so I presume it's all usb-3. My
Crosshair-IV Formula mobo has 2 usb3 and 6 usb3 ports, both types are
enabl;ed in BIOS. If I plug the gx-100 into a blue usb-3 port then
lsusb will NOT show it. Go figure. If I plug it into a usb-2 port then
it shows up.
The vBox host usb setup is a bit crooked but if I select usb-2 then it
will autodetect the device and auto-fill the the id numbers. Here the
device should NOT show up in the filter with usb-3 seleted because the
host cannot see it then. Anyway, I select usb-2 and the filter is
enabled though I don't see the point of this filter at all, all detected
usb devices show up under the guest window regardless of what's in this
filter, including the gx-100 if usb-2 is selected in the host.
If I set 3 in host then the guest w10 Device-Manager last entry also
shows USB 3, if I set 2 in the host then Device-Manager shows 2. The
very few times I got it to work I never had to tinker in the guest
Device-Manager. If I set usb-2 in the host, filter or no filter, then
the gx-100 shows up in the strip just below the w10 window BUT GHOSTED
and the bubble for it shows all the right numbers plus that it is
"unavailable".

Is the extension pack installed in the host? If that's been answered before,
I've forgotten.
https://download.virtualbox.org/virtualbox/7.0.14/Oracle_VM_VirtualBox_Extension_Pack-7.0.14.vbox-extpack

Yes, sometimes even twice to be sure :-)

When apache acts up you can look into its error logs, does
vbox have anything like it?

Paul

2024-02-24 06:06:48 UTC

Double check the settings for the guest to ensure it has usb 2 or 3 enabled
matching what the host sees. Usb 2 is for ohci/ehci, while 3 is for xhci.

The gx-100 was made in 2022-2023 so I presume it's all usb-3. My Crosshair-IV Formula mobo has 2 usb3 and 6 usb3 ports, both types are enabl;ed in BIOS. If I plug the gx-100 into a blue usb-3 port then lsusb will NOT show it. Go figure. If I plug it into a usb-2 port then it shows up.
The vBox host usb setup is a bit crooked but if I select usb-2 then it will autodetect the device and auto-fill the the id numbers. Here the device should NOT show up in the filter with usb-3 seleted because the host cannot see it then. Anyway, I select usb-2 and the filter is enabled though I don't see the point of this filter at all, all detected usb devices show up under the guest window regardless of what's in this filter, including the gx-100 if usb-2 is selected in the host.
If I set 3 in host then the guest w10 Device-Manager last entry also shows USB 3, if I set 2 in the host then Device-Manager shows 2. The very few times I got it to work I never had to tinker in the guest Device-Manager. If I set usb-2 in the host, filter or no filter, then the gx-100 shows up in the strip just below the w10 window BUT GHOSTED and the bubble for it shows all the right numbers plus that it is "unavailable".

Not every peripheral is USB3. Even if made today, there are still lots
of USB2 peripherals.

USB3 has 9 contacts. A row of 4. A row of 5.

The row of five, is a differential high speed interface.
This is similar to how PCI Express works or SATA works. The
ground in the center, would be for crosstalk. The SATA one
has three GND, with the seven pin interfacing having GND
on the outside of the set of five in my picture.

TX+ TX- GND RX+ RX-

Those are low amplitude signals, like 1 volt amplitude.

Versus USB2 which is a 5V logic. VBUS is 5V. The D+ and D-
are higher amplitude signals, and run half-duplex (either
transmit, or receive, but not both).

VBUS D+ D- GND

You'll notice the USB3 five pins, cannot work without a power
source, and VBUS is the power source for the logic (even if the
logic is USB3).

*******

Motherboards come two ways.

If a motherboard is manufactured today, virtually all the logic
blocks for USB, are USB3 blocks and XHCI driver standard. To make
a USB2 connector on such motherboards, they just connect the four
pins of a USB3 block. This then, becomes USB2 with XHCI driver.

Older motherboards (ten years ago), the USB2 ports were conventional
OHCI/EHCI, while the USB3 ports were XHCI and sometimes produced by
an add-on chip, instead of via the Southbridge. Such ports are
"genuine articles", no screwing around. And the USB2 "should just work"
on a ten year old motherboard.

To add genuine USB2 to a motherboard, via add-on card, is tough now.
New motherboards typically do not have PCI slots. The ten year
old machine has a PCI slot. Genuine USB2 addon cards (like a NEC chip)
were PCI chips. Rather than PCI Express ones.

While you could connect a USB2 to PCI, to a PCI to PCIe bridge chip,
companies stopped doing that, after the bridge company got bought
out, and the pirate ship that bought them, raised the prices. This
immediately sunk the add-on legacy card business for PCIe. It was
on thin ice to begin with, and raising the bridge price, everyone
just exited the market. At the same time, motherboards stopped
being made with four bifurcation chips between two x16 connectors
for PCI Express, as back then, you could have x16, x0 or x8, x8
for motherboard wiring, and four DIP addon chips handled the routing.
The pirate company raised the price on those too, so those had
to go out the window. Now, motherboards have an x16 slot and an x4
slot and zero bifurcation chips.

*******

In VirtualBox, the logs are in the storage area for the VM container,
in a folder called "Logs". So if I had a Ubuntu VM and a Mint VM,
then the Ubuntu folder has a "VBox.log" in its Logs folder and
the Mint folder has a "VBox.log" in its Logs folder.

I sometimes move the entire folder for storage elsewhere. Which
means I am responsible for finding the VBox.log when I need it.
(It's inside a ZIP I make of the folder.)

It's worth a look, that log, but no guarantees as to what is going on.

Paul

bad sector

2024-02-24 12:36:33 UTC

Double check the settings for the guest to ensure it has usb 2 or 3 enabled
matching what the host sees. Usb 2 is for ohci/ehci, while 3 is for xhci.

The gx-100 was made in 2022-2023 so I presume it's all usb-3. My Crosshair-IV Formula mobo has 2 usb3 and 6 usb3 ports, both types are enabl;ed in BIOS. If I plug the gx-100 into a blue usb-3 port then lsusb will NOT show it. Go figure. If I plug it into a usb-2 port then it shows up.
The vBox host usb setup is a bit crooked but if I select usb-2 then it will autodetect the device and auto-fill the the id numbers. Here the device should NOT show up in the filter with usb-3 seleted because the host cannot see it then. Anyway, I select usb-2 and the filter is enabled though I don't see the point of this filter at all, all detected usb devices show up under the guest window regardless of what's in this filter, including the gx-100 if usb-2 is selected in the host.
If I set 3 in host then the guest w10 Device-Manager last entry also shows USB 3, if I set 2 in the host then Device-Manager shows 2. The very few times I got it to work I never had to tinker in the guest Device-Manager. If I set usb-2 in the host, filter or no filter, then the gx-100 shows up in the strip just below the w10 window BUT GHOSTED and the bubble for it shows all the right numbers plus that it is "unavailable".

Not every peripheral is USB3. Even if made today, there are still lots
of USB2 peripherals.
USB3 has 9 contacts. A row of 4. A row of 5.
The row of five, is a differential high speed interface.
This is similar to how PCI Express works or SATA works. The
ground in the center, would be for crosstalk. The SATA one
has three GND, with the seven pin interfacing having GND
on the outside of the set of five in my picture.
TX+ TX- GND RX+ RX-
Those are low amplitude signals, like 1 volt amplitude.
Versus USB2 which is a 5V logic. VBUS is 5V. The D+ and D-
are higher amplitude signals, and run half-duplex (either
transmit, or receive, but not both).
VBUS D+ D- GND
You'll notice the USB3 five pins, cannot work without a power
source, and VBUS is the power source for the logic (even if the
logic is USB3).
*******
Motherboards come two ways.
If a motherboard is manufactured today, virtually all the logic
blocks for USB, are USB3 blocks and XHCI driver standard. To make
a USB2 connector on such motherboards, they just connect the four
pins of a USB3 block. This then, becomes USB2 with XHCI driver.
Older motherboards (ten years ago), the USB2 ports were conventional
OHCI/EHCI, while the USB3 ports were XHCI and sometimes produced by
an add-on chip, instead of via the Southbridge. Such ports are
"genuine articles", no screwing around. And the USB2 "should just work"
on a ten year old motherboard.
To add genuine USB2 to a motherboard, via add-on card, is tough now.
New motherboards typically do not have PCI slots. The ten year
old machine has a PCI slot. Genuine USB2 addon cards (like a NEC chip)
were PCI chips. Rather than PCI Express ones.
While you could connect a USB2 to PCI, to a PCI to PCIe bridge chip,
companies stopped doing that, after the bridge company got bought
out, and the pirate ship that bought them, raised the prices. This
immediately sunk the add-on legacy card business for PCIe. It was
on thin ice to begin with, and raising the bridge price, everyone
just exited the market. At the same time, motherboards stopped
being made with four bifurcation chips between two x16 connectors
for PCI Express, as back then, you could have x16, x0 or x8, x8
for motherboard wiring, and four DIP addon chips handled the routing.
The pirate company raised the price on those too, so those had
to go out the window. Now, motherboards have an x16 slot and an x4
slot and zero bifurcation chips.
*******
In VirtualBox, the logs are in the storage area for the VM container,
in a folder called "Logs". So if I had a Ubuntu VM and a Mint VM,
then the Ubuntu folder has a "VBox.log" in its Logs folder and
the Mint folder has a "VBox.log" in its Logs folder.
I sometimes move the entire folder for storage elsewhere. Which
means I am responsible for finding the VBox.log when I need it.
(It's inside a ZIP I make of the folder.)
It's worth a look, that log, but no guarantees as to what is going on.
Paul

I'll see if i can find it tonight, this usb whorehouse invarably reminds
me of my durgod (otherwise great!) keyboard. It's just about impossible
to get into BIOS with it. With the gx-100 and w10 I'll test on a fresh
tumbleweed install with vbox installed right in the net-install chain
with a live dvd and only a single user in the only additional (vboxuser)
group.

bad sector

2024-02-25 03:29:13 UTC

"Unknown USB Device, Device Descriptor Request Failed"

I don't have a realwin installed at the moment so I'll
have to bring presents (usually in cases of 12/24) to
someone who does, to test further.

Paul

2024-02-25 13:41:58 UTC

I got one more yard downfield but still no cigar.
Installed a freshie Tumbleweed with only one user
who is in group vboxusers. This user's home (and
the vbox belonging to) are now both on the same
partition instead of elsewhere with/without linking.
The gx-100 usb connection in the strip below the
w10 guest window is no longer ghosted so I would
assume that the passthrough is good.
BUT... now winblows is acting up with the annunciations
https://imgur.com/dYm1bf2.png
"Unknown USB Device, Device Descriptor Request Failed"
I don't have a realwin installed at the moment so I'll
have to bring presents (usually in cases of 12/24) to
someone who does, to test further.

I don't think you can get around that, by unplugging and
plugging back in. As the Host gets first crack at it,
and only then does the Guest filter transfer control
to the Guest OS. This means whatever foul-up leaves it in
that state, is reproducible.

The way that VirtualBox works, is it is a "USB packet
redirector". It does not work at the Class level.
Every packet for 1234:5678 should be redirected to the Guest.
And that would include responses to a Device Descriptor Request.

first you make endpoints, and collecting the config space info
comes next. And that's what the Device Descriptor is.

These are things you can inspect (the Device Descriptor),
in USBTreeView in Windows. But the complex table shown
as Device Descriptor, is only going to show... if it hasn't
failed. Like, on a friends Windows PC perhaps. Take your
USBTreeView with you, so you can take reference pictures
of what the device entry *should* look like.

There is probably a "STOP code" in Device Manager for the
affected device, but going down that rabbit hole isn't going
to help, as this is more likely to be a VirtualBox side
effect than a "pure" Windows problem.

Paul

bad sector

2024-02-28 00:13:56 UTC

I don't think you can get around that, by unplugging and
plugging back in. As the Host gets first crack at it,
and only then does the Guest filter transfer control
to the Guest OS. This means whatever foul-up leaves it in
that state, is reproducible.
The way that VirtualBox works, is it is a "USB packet
redirector". It does not work at the Class level.
Every packet for 1234:5678 should be redirected to the Guest.
And that would include responses to a Device Descriptor Request.
first you make endpoints, and collecting the config space info
comes next. And that's what the Device Descriptor is.
These are things you can inspect (the Device Descriptor),
in USBTreeView in Windows. But the complex table shown
as Device Descriptor, is only going to show... if it hasn't
failed. Like, on a friends Windows PC perhaps. Take your
USBTreeView with you, so you can take reference pictures
of what the device entry *should* look like.
There is probably a "STOP code" in Device Manager for the
affected device, but going down that rabbit hole isn't going
to help, as this is more likely to be a VirtualBox side
effect than a "pure" Windows problem.

I recovered my laptop to its w10 delivered state and DL'd this TS kit
including "usbview" (the gx-100 BTW works like a charm in w10 so there
are NO hardware issues). But the flood of info is a chinese jungle to
me; which of these should the GX-100 usb entry in virtualbox repeat and
how can one enforce that?

https://transfiles.ru/yy63w

(it's up for 3 days)

Paul

2024-02-28 02:09:31 UTC

I don't think you can get around that, by unplugging and
plugging back in. As the Host gets first crack at it,
and only then does the Guest filter transfer control
to the Guest OS. This means whatever foul-up leaves it in
that state, is reproducible.
The way that VirtualBox works, is it is a "USB packet
redirector". It does not work at the Class level.
Every packet for 1234:5678 should be redirected to the Guest.
And that would include responses to a Device Descriptor Request.
first you make endpoints, and collecting the config space info
comes next. And that's what the Device Descriptor is.
These are things you can inspect (the Device Descriptor),
in USBTreeView in Windows. But the complex table shown
as Device Descriptor, is only going to show... if it hasn't
failed. Like, on a friends Windows PC perhaps. Take your
USBTreeView with you, so you can take reference pictures
of what the device entry *should* look like.
There is probably a "STOP code" in Device Manager for the
affected device, but going down that rabbit hole isn't going
to help, as this is more likely to be a VirtualBox side
effect than a "pure" Windows problem.

I recovered my laptop to its w10 delivered state and DL'd this TS kit including "usbview" (the gx-100 BTW works like a charm in w10 so there are NO hardware issues). But the flood of info is a chinese jungle to me; which of these should the GX-100 usb entry in virtualbox repeat and how can one enforce that?
https://transfiles.ru/yy63w
(it's up for 3 days)

OK, it appears to have made endpoints in usbview.txt .

But this says it all. It does not follow a
USB Class specification (and that's OK because custom
devices have always existed in USB).

bDeviceClass: 0xFF -> This is a Vendor Specific Device

They're not as common as it used to be, because a
lot of custom ones have become standardized.

So the question would be, why is VirtualBox getting bent
out of shape about a custom ? It still has to bind to a driver
via its PNP numbers.

I would guess it's failing earlier than when a report like
that can be made. It's having a problem making endpoints,
at a guess.

Is it on a real USB2 port or is it using the four contact
part of a nine contact USB3 port ? I don't think the machine
I'm typing on, as an example, has "real" USB2 ports. The ports
with black tabs in the connector, those are made from
four of nine contacts on a USB3 connector.

I don't really think it is the USB port that deserves
the blame. But I don't know how to fix this.

*******

What your report does give us, is a Google search: "bDeviceClass: 0xFF virtualbox"

While a lot of the hits are problem reports, this one is a bit different bInterfaceClass 255 Vendor Specific Class

https://kicherer.org/joomla/index.php/de/blog/38-reverse-engineering-a-usb-sound-card-with-midi-interface-for-linux

I always thought that setting up endpoints, that's a hub driver or similar, function.
It's not up to a device driver to do that. The declaration in the device, should
contain anything it needs to know about endpoints. That's part of what
makes it hard to understand why that is failing.

Paul

bad sector

2024-03-03 20:14:06 UTC